quasar-ai 1.1.0

🚀 QUASAR - AI-powered CLI code editor with agentic capabilities

🚀 QUASAR - AI-Powered CLI Code Editor

An intelligent command-line assistant that can understand your codebase, generate code, fix bugs, and execute tasks using AI.

Installation

pip install quasar-ai

Setup API Keys

IMPORTANT: You must provide your own API keys. QUASAR does not include any API keys.

Option 1: Using .env file (Recommended)

Create a .env file in your project directory:

# Groq (recommended - fast inference)
GROQ_API_KEY_1=gsk_your_key_here
GROQ_API_KEY_2=gsk_your_second_key_here

# Cerebras
CEREBRAS_API_KEY_1=csk_your_key_here

# Ollama runs locally - no API key needed

Option 2: Using Environment Variables

# Groq
export GROQ_API_KEY_1="gsk_your_key_here"
export GROQ_API_KEY_2="gsk_your_second_key_here"

# Cerebras
export CEREBRAS_API_KEY_1="csk_your_key_here"

Multiple Keys & Fallback Behavior

You can add multiple keys per provider (e.g., GROQ_API_KEY_1, GROQ_API_KEY_2, GROQ_API_KEY_3).

In Auto mode (default):

• If the first key hits rate limits or fails, QUASAR automatically tries the second key
• If all keys for a provider fail, it falls back to the next provider
• Fallback chain: Groq → Cerebras → Ollama

Get free API keys:

• Groq: https://console.groq.com
• Cerebras: https://cloud.cerebras.ai

Usage

Interactive Mode (REPL)

quasar
# or
quasar --interactive

Single Command

quasar "create a hello.py file that prints Hello World"
quasar "explain main.py"
quasar "fix the bug in utils.py"
quasar "list files in current directory"

Specify Workspace

quasar --workspace /path/to/project "add tests for api.py"

Custom Model Selection

By default, QUASAR automatically selects the best model for each task. You can override this with --model:

# Use a specific Cerebras model
quasar --model cerebras/qwen-3-32b "explain this code"

# Use Groq with a specific model
quasar --model groq/llama-3.3-70b-versatile "create a REST API"

# Use local Ollama model
quasar --model ollama/qwen2.5-coder:7b "fix the bug"

# Interactive mode with custom model
quasar -i -m cerebras/qwen-3-32b

Note: When you select a model, it will be used for ALL tasks. Choose a model that supports tool calling and has good reasoning capabilities.

Supported Tasks

QUASAR automatically classifies your request and uses the best model:

Task	Example
Chat	"What is machine learning?"
Code Generation	"Create a REST API endpoint"
Bug Fixing	"Fix the TypeError in app.py"
Code Explanation	"Explain this function"
Refactoring	"Improve the structure of utils.py"
Documentation	"Add docstrings to main.py"
Test Generation	"Write tests for calculator.py"

Web Tools (Beta)

QUASAR can search the web and read URLs to help with your tasks:

quasar "search for the latest Python best practices"
quasar "read the documentation at https://docs.python.org/3/library/asyncio.html"

Web Search Configuration

Add to your .env file:

# Tavily API Key - Get from https://tavily.com
TAVILY_API_KEY=your_tavily_api_key_here

# SearXNG Host (if self-hosting)
# SEARX_HOST=http://localhost:8080

⚠️ Beta: Web tools are in beta phase. Results may vary.

Updating

pip install --upgrade quasar-ai